Datasets:

hndc

/

AI4Hazard-small

like 0

166 Human Factors Handbook and Risk Index does not prescribe acceptab le levels of fatigue risk. It advocates that fatigue and risk scores should be reduced as a low as reasonably practicable. The Fatigue and Risk Index also gives a risk score that indicates the potential for an accident/incident to occur. In all examples it was assumed that th e person takes 40-minutes to travel to work and has a “moderately” demanding role that required attention “most of the time”. A long day Figure 15-1 shows an example of how fatigue “jumps” when working a 16-hour day. The fatigue score rises slowly from da y one to day four. Then the fatigue score jumps on day five. It then stays high for days six and seven. Figure 15-1: Example of rapid rise in fatigue scores from a 16-hour day Working without rest breaks Figure 15-2 shows the impact of working 12-hour day shifts without rest breaks. The “no rest break” example assumes a 15 -minute lunch break after four hours, and no other breaks. The “with rest breaks” example assumes a 15-minute break every 2.5 hours and a half hour lunch break. The lack of rest breaks causes fatigue scores to be increased three times. The fatigue level at the end of the seven days without rest breaks would be roughly a one in three chance of struggling to stay awake at work.

E.17 Not Invented Here |303 for their potential acceptance/reaction by managem ent. They deleted any recommendation thought to be too expensive, time- consuming, or difficult. Occasionally, the risk rankings were re- assigned so that recomm endations not be necessary. During an audit, interviews with som e of the team leaders revealed that they believed that it was their responsibility to m ake the recommendations addressing problems identified in the PHA go away. When pressed further about why not m ake the problems go away by truly addressing them, each responded “There’s no energy for that here.” The team leaders believed management did not want to be the ones to decide not to address a recomm endation. Some believed that their performance would be adversely evaluated if they submitted PHA reports with major recomm endations. In several cases, PHA’s were re-convened to revise the risk rankings and recomm endations to make them less onerous or unnecessary. Who has the responsibility to choose between implementing recomm endations or accepting risk? Establish an Imperative for Safety, Understand and Act Upon Hazards/Risks, Provide Strong Leadership. E.17 N ot Invented Here A new PSMS Coordinator attempted to incorporate several good practices from the facility where he previously worked. He believed the facility could benefit from these ideas and that they would be a relatively good fit with his new site’s PSM S, personnel, and policies. His manager disagreed, saying that the Coordinator’s previous com pany was different, the practices were actually poor fits, and they would be too time-consum ing and upsetting to implement som ething different when the current PSMS seemed to be running sm oothly. B ased on Real Situations

195 making the abandoned equipment unusable if it cannot be cleaned adequately. Example 8.2 Reactors equipped with heavy agit ators used for tetraethyl lead manufacture during World War II we re disinterred from bomb rubble and were found by the people who dug them up to be ideal for processing fish paste for human consumption. The reactors were washed, but this did not prevent poisoning a number of people. There is a temptation for manage rs to delay the dismantling of decommissioned or abandoned plants as long as possible, usually due to the costs of demolition. However, experience teaches that there will never be a time for chemical plant cl osure activities that will be less expensive or less hazardous than immediately after the plant is closed, because it is the time when: the most people who know how to handle the materials and where the materials are located are available. units are still intact, decontamination can be most easily performed, the procedures are well-known, and the equipment is available. waste disposal contracts that co ver the materials in the plant are still open or can be re-opened easily. design documents, waste mani fests, maintenance records, and other files are most likely to be readily available. equipment has probably not corroded to the point that it can’t be handled safely—valves will open, nuts are not frozen, and instruments are in working condition. you are least likely to encounter tanks, drums, etc., with contents nobody can identify without an expensive analytical investigation. 8.10 TRANSPORTATION Transportation of chemic als within a facility and to/from a facility is not a distinct phase in the life cycle of a chemical process, as it is a vital support activity that occurs consta ntly, in most facilities every day.

22. Human Factors in emergencies 279 Figure 22-1: Error recognition and management process (Adapted from [88]) Emergency situations increase the lik elihood of human errors. Figure 22-2 shows six main categories of errors [89] . These are errors of action, checking, retrieval, transmission, diagnosis, and decision. Emergency scenario incidents commonly present with multiple human errors. Error Recognition What is the problem? How much time is available? How risky is the situation (present and future)? Time Limited High Risk Time Available Risk Variable Problem Understanding YES NO Evaluate options Execute actions Review outcomes Gather more Information Put the process into a safe state: shut down

262 PROCESS SAFETY FOR ENGINEERS: AN INTRODUCTION References AESolutions, AEShield, https://www.aeshield.com. CCPS Glossary, “CCPS Process Safety Glossary ”, Center for Chemical Process Safety, https://www.aiche.org/ccps/resources/glossary . CCPS 1999, Guidelines for Chemical Proce ss Quantitative Risk Analysis , Center for Chemical Process Safety, John Wiley & Sons, Hoboken, N.J. CCPS 2001, Guidelines for Revalidating Process Hazard Analysis, 1st edition , Center for Chemical Process Safety, John Wiley & Sons, Hoboken, N.J. CCPS 2008 a, Incidents That Define Process Safety , Center for Chemical Process Safety, John Wiley & Sons, Hoboken, N.J. CCPS 2008 b, Guidelines for Hazard Evaluation Procedures , Center for Chemical Process Safety, John Wiley & Sons, Hoboken, N.J. CCPS 2012, Guidelines for Engineering Design for Process Safety , Center for Chemical Process Safety, John Wiley & Sons, Hoboken, N.J. CCPS 2013, Guidelines for Managing Process Safety Risks During Organizational Change , Center for Chemical Process Safety, Jo hn Wiley & Sons, Hoboken, N.J. CCSP 2018, Guidelines on Integrating Process Safety into Engineering Projects , Center for Chemical Process Safety, John Wiley & Sons, Hoboken, N.J. Crawley, Crawley, F. and Tyler, B., HAZOP: Guide to Best Practice, 3rd Edition , Institution of Chemical Engineers, Elsevier, U.K., 2015. Crowl 2019, Daniel A. Crowl and Joseph F. Louvar, Chemical Process Safety, Fundamentals with Applications 4th Edition ., Pearson, NY. Enggcyclopedia, www.enggcyclopedia.com/2012/05/hazop-study/ EFCE 1985, “Risk Analysis in the Process Indust ries, Report of the International Study Group on Risk Analysis”, European Federation of Chemical Engineering, EFCE Publications, Series No. 45, Institute of Chemical Engineers, Rugby, England. LRC. “The Esso Longford Gas Plant Accident Report of the Longford Royal Commission”, Government Printer for the State of Victoria, No. 61 – Session 1998 -99. June 28,1999, https://www.parliament.vic.gov.au/ papers/govpub/VPARL1998-99No61.pdf OSHA, Hazard Communication, 29 CFR 1910. 1200, Occupational Safety and Health Administration, https://www.osha.gov/laws- regs/regulations/standardnumber/1910/1910.1200 Primatech, PHAWorks https://www.primatech.com/software/phaworks Sphera, https://sphera.com/pha-pro-software/

DETERM INING ROOT CAUSES 245 10.7.3 Predefined Tree The causal factors need to be examined further to determine why those factors existed. The investigation team may use a predefined tree to examine each causal factor individually. Th e first causal factor is analyzed, starting at the top of the tree and then wo rking down all of the branches as far as the facts permit. When an appropri ate subcategory on one of the branches is identified, it is recorded as a root cause. The remain ing branches are checked, as one causal factor may have multiple root causes. The procedure is then repeated for each causal factor in turn. Several quality assurance tests should be applied when using predefined trees. This is an import ant step because predefin ed trees are designed to capture most root causes, but they may not be comprehensive. A completeness check should be conducted on each branch of the tree to see if there are other root causes associated with the category of that branch that are not list ed on the tree. Some predefined trees do fully reach down to the root cause level. A system test should be applied to each identified root cause to ensure that it relates to a management system failure. By applying the 5-Whys tool to each cause identified at the end of the re levant branches of the tree, the investigator can determine if anothe r underlying cause can be identified. After the predefined tree has been used, a final generic cause test should be applied. The plant operating histor y, especially previous incidents, is considered to indicate if other generic management system problems exist. For example, repetitive failures may in dicate generic causes that would not be apparent by only investigating the current incident. It is also an opportunity for a final overall review of the investigation to focus on the big picture, not just individual facts or causal factors. The team should ask, “Are there any other causes that anyone has in mind that have not been included?” If the incident investigation team is sati sfied with the root causes identified, then the investigation proceeds to the recommendation stage. If a problem or some incompleteness is noted, th en an iterative loop is followed.

ACKNOWLEDGMENTS xxxiii Before publication, all CCPS books are subjected to a thorough peer review process. CCPS gratefully ac knowledges the thoughtful comments and suggestions of the pe er reviewers. Their work enhanced the accuracy and clarity of these guidelines. Althou gh the peer review ers have provided many constructive comments and sugg estions, they were not asked to endorse this book and were not shown the final manuscript before its release. Peer Reviewers: Mohd Ibrahim Mohd Ashraf Petronas Palaniappan Chidambaram DuPont Sustainable Solutions Robert Johnson Unwin Company Caitlin Mullan Ashland Connor Murray CNRL Beverley Perozzo NOVA Chemicals

234 Human Factors Handbook 19.2.2 Contributing Human Factors This was a very complex event with a series of equipment failures, errors, and mistakes. It escalated over about five hours. There was a high volume of communication between dispersed personne l by radio. They were communicating safety critical information. As the event escalated, actions should have been performed quickly. There were omissions in the verbal communication of plant state, including concerning levels in an absorber, and between the preceding night shift and the day shift. There were also omissions in a log (reasons for not operating a locked and tagged bypass valve). The verbal communication technique was weak and prone to error. There was a simple mishearing of “P” instead of “T” over the radio. The production coordinator heard what he expected to hear – a “typical” human error. 19.3 Causes of poor communication People in process plants often communi cate with one another from different locations, such as from a control room to a remote part of the site. People may be working in an area with high levels of noise from equipment and may be wearing hearing protection. Words and sentences can be partly obscured by “radio noise” or weak signals. The recipient may mishear what is being said or incorrectly “fill in” the missing words. People have a limited cognitive capacity for receiving information. Information (including verbal communication) is filtered to avoid overload. This is known as “selective attention”. Selective attention makes it possible to think and make decisions without being overloaded. It also creates a risk of unintentionally and unknowingly filtering out safety critical in formation. In addition, as noted in 16.3, people may be focused on a task or multitasking. This creates a risk that audible verbal communication is not registered by the recipient. If the amount of information being communicated exceeds the short-term memory, then some of the information will be forgotten. It is often necessary in a proce ss environment to communicate precise information, such as a valve number, or give warnings or instructions for safety critical tasks. If words, letters, or numbers are used that sound similar to other words, letters, or numbers, the recipient is more likely to hear something different to what was said. For example, the letters “D” and “P” are easily confused, as are A complex safety critical operation involving a high volume of communication requires a high reliability communication process.

110 | 4 Applying the Core Pr inciples of Process Safety Culture In general, when designing an incentive system for process safety, the following points should be considered. Consider the Potential for Inverse Effects A company goal to reduce the num ber of incidents from year- to-year is certainly desirable. However, using incident number or incident reduction for purposes of incentive may drive personnel to hide or under-report incidents. Whatever basis for incentive is considered, leaders should think about how it could lead to the opposite of the desired behavior. It may also m ake sense to independently validate incentive metrics to ensure this has not happened. Focus on the Frequent, not the Rare Since the ultimate goal is to prevent process safety incidents, it can be tempting to use the lagging process safety incident rate as the basis for incentive. The problem is that incident rates are generally low, and a leader can perform poorly in process safety for a long time before an incident occurs. It is better to avoid using lagging m etrics, such as incident rate. Instead use leading metrics related to correct behaviors that m ust happen frequently over time such as percent com pletion of asset integrity actions (e.g. inspection, testing, and preventative maintenance). Near-misses occur m uch more frequently and can also be an option. Focus on the Long-Term , not the Short-Term Since process safety needs to be performed well, consistently over time, the basis for incentives should consider long-term perform ance. This can be easier to accomplish in incentive schem es that have a multi-year basis, but still is possible in year- by-year schem es. For example, the incentive should consider whether the goal was intended to have been reached by steady perform ance over the year, and penalize individuals who achieved

140 INVESTIGATING PROCESS SAFETY INCIDENTS Obtain approval from authority havi ng jurisdiction over the scene Prevent loss of or damage to eviden ce ("spoliation of evidence"). Under certain circumstances, and usin g strict control measures, it may be helpful to allow duplication of paper or electronic records and/ or material samples and make these available to th e other stakeholders. Agreements can also be reached regarding mutually acceptable testing laboratories and other outside resources when limited quantities or unique pieces of evidence necessitate that all interested partie s cooperate in evidence analysis. The investigation team may be faced with the challenge of determining what equipment was the source of an explosion and what was damaged as a result of an explosio n. Fragments and debris can be thrown considerable distances, sometimes outside facility boundaries. Loss of plant utilities, chemical spills, and si gnificant blast damage to adjacent process units and buildings may greatly hamper the invest igation or even prohibit access to the site for days or longer. Identifying and capturing time-sensitive evidence is the top priority at the outset of an investigation to limit the potential for evidence deterioration due to exposure and loss of plant utilities. Electronic process data, chemical samples, fragments outside of facility boundaries, and evidence that may be altered by emergency responders and HAZMAT teams are typically high priority and should be gathered quickly. The loss of electric power to control systems places urgency on the collection of electr onic data since battery backups have a limited lifespan, sometimes measured in hours or less. Chemical feed and product sa mples should be obtained from the area if possible since the material ac tually in process may have been consumed or ejected during the explosion. Fragments thrown beyond facility boundaries may be picked up by untrained individuals, and may not be returned to the plant. Offsite damage is also beyond company control, and documentation of the extent of damage may be necessary on an expedient basis, before repairs are made. Evidence that is less time sensitive and within facility boundaries is second priority to collect. Plant person nel can better control such evidence. Nonetheless, evidence may be spread over a large area, and all personnel within the plant must be instructed on the proper manner to communicate the location of evidence for co llection by a trained team.

CASE STUDIES/LESSONS LEARNED 167 increasingly available for training purposes. The aviation industry has been highly regulated for many years, although it has recently started to introduce formal safety management systems (SMSs) that provide a top- down, organization-wide approach to managing safety. An “Advisory Circular” was issued to the US aviation industry in January 2015 (USDoT Advisory Circular) requiring service providers to develop SMSs and was followed by 14-CFR 119.8 Safety Mana gement Systems, which requires an SMS to be in place by March 19, 2018 (ECFR Title 14, Aeronautics and Space ). In Europe, the Civil Aviation Authority provided CAA CAP 795: Safety Management Systems (SMS ) guidance for organisations in 2015 (CAA 2015). Formal safety management systems have been in place in the process industries for many more year s than in the aviation sector. For designated highly hazardous chem icals in the US, this was first mandated by the Process Safety Management (PSM) regulations 29CFR 1910.119 (OSHA PSM) in 1992. In 1 984, the UK implemented the CIMAH (Control of Industrial Major Accide nt Hazards) Regulations; and in Europe, the ‘Seveso Directive’, on th e control of major accident hazards involving dangerous substances, wa s originally published in 1996 and became law in 1999. In the UK, the Seveso Directive replaced CIMAH and the UK adopted it as COMAH (Control of Major Accident Hazards) in 1999. The CCPS produced their text Guidelines for Risk Based Process Safety in 2007 and the Energy Institute produced a High Level Framework for Process Safety Management in 2010 (CCPS 2007, EI 2010). Despite the difference in the timing of implementing formal systems for safety management, many of the issues and features of the modern- day cockpit can involve challenges sim ilar to those in process industry control rooms when it comes to ad dressing abnormal situations. These include, but are not limited to: Information and alarm overload. Increased reliance on automation. Less opportunity to practice reacting to abnormal situations. The “startle effect”, when an automated system suddenly cannot control the process, an d the operator has to take rapid action.

Piping and Instrumentation Diagram Development 132 However, we know (based on the concepts mentioned in Chapter 5) that when a process item is taken out of operation, consideration should be made of the way the plant (or unit) should operate in the absence of that ele-ment. There are some cases where there is no techni-cally acceptable solution that effectively mitigates the lack of item so that the rest of plant can operate with minimum impact. In such cases one may essentially challenge the need for placing an isolation system for that specific item. One example is heat exchangers. When you put a heat exchanger on a stream in your plant you want to increase or decrease the temperature (and maybe the phase) of your stream. Most likely that temperate change is so important that you were forced to put in a heat exchanger. Now, what would you do in the absence of a heat exchanger? If you review the avail-able options mentioned in Chapter 5, you may see it is very hard to find an attractive solution. If this is the case, why you should bother to put in an isolation system “for the time you need to pull the heat exchanger out of operation”? Therefore the other question you need to ask yourself before placing an isolation system is whether you can “afford” to be without that piece of equipment in the plant or not? This is the reason that some companies don’t provide isolation systems for their heat exchangers apart from a few exceptions. Exceptions could be when there are spare heat exchangers available in parallel, when there is an automatic cleaning system for the heat exchanger that can clean it in a short time, and/or when the target stream leaving the heat exchanger goes into a large container. In the first case we obviously need an isolation system to bring the spare heat exchanger into service. In the second and third cases, we again need an isolation system, but we do nothing during the time we are lacking the heat exchanger; however, we still can afford it because it is a short time and/or the disrupted stream (with non‐suitable temperature) goes into a large container and is thermally equalized with the bulk fluid in it in a way that only a small and acceptable temperature change can be observed. 8.7.2 Type of Isola tion System The second question is: what is the isolation arrange- ment? To answer this question, the concept of isolation should be discussed. Isolation in this context means the segregation of a piece of equipment, or even a portion of the plant, from the rest of the plant while the plant is operating. Isolation is done using an “isolation system. ” The general concept of an isolation system is shown in Figure 8.2. The purpose of isolation could be inspection, cleaning, in‐place repair, workshop maintenance, etc. One may say that isolation can be provided simply by closing the inlet and outlet valves. You can see such an arrangement for a pressure gauge in Figure 8.3. The valve symbol shows a valve that we generally call a “root valve” because it is installed at the root of the pressure gauge. This root valve serves as an “isolation valve” and is a ball valve or gate valve. If someone needs to inspect and/or re‐calibrate the pressure gauge, they can do it without emptying the vessel. Although a single simple valve can be accepted as an isolation system in some non‐complicated systems (like instruments), it is generally not accepted for the isolation of equipment. “Root valves” are a type of isolation method used for instruments. tm = period betw een each maintenance/inspection/cleaning tm = 2 yrtm = 2 yr tm = 2 yr tm = 2 yr tm = 2 yrtm = 1 yr Figure 8.1 Dependenc y of need or lack of need for isolation systems for items. PipeIsolation systemIsolation systemPiece that needs to be isolatedFigure 8.2 Gener al overview of isolation. PG 103 Figure 8.3 Root v alve for isolation of a pressure gauge.

328 PROCESS SAFETY FOR ENGINEERS: AN INTRODUCTION consequence pairs for evaluation, such as a HAZOP. In a HAZOP, safeguards are identified. In LOPA, the independent protection layers (IPL) ar e identified from the lists of safeguards and only the IPLs are credited in reducing the risk . Safeguards that are no t usually considered to be IPLs include: training, proced ures, maintenance, and signage. Figure 14.12. Typical layers of protection (redrawn from CCPS 2001) Independent Protection Layer (IPL) - A device, system, or action that is capable of preventing a scenario from proceeding to the undesired consequence without being adversely affected by the initiating event or the action of any other protection layer associated with the scenario. Note: Protection layers that are designated as "independent” must meet specific functional criteria. A protec tion layer meets the requirements of being an IPL when it is designed and managed to achieve the following seven core attributes: Independent; Functional; Integrity; Reliable; Validated, Maintained and Audited; Access Security; and Management of Change. (CCPS Glossary) In order to be considered an IPL, a device, system, or action must be: effective in preventing the consequence when it functions as designed, independent of the initiating event and the components of any other IPL already claimed for the same scenario, and auditable; the assumed effectiveness in terms of consequence prevention and probability of failure on demand (PFD) mu st be capable of validation in some manner (by documentation, review, testing, etc.). The following is an example of how the spectr um of tools may be used in supporting risk- based decision making.

APPLICATION OF PROCESS SAFETY TO WELLS 81 Figure 4-5. Example showin g different levels of detail in bow tie diagrams Two useful definitions of well integrity are as follows. API 100-1 (2015b): “The quality or condition of a well in being structurally sound and with competent pr essure seals (barriers) by application of technical, operational, and organizational solutions that reduce the risk of unintended subsurface movement or uncontrolled release of formation fluid.” NORSOK D-010 (2013): “Application of technical, operational and organizational solutions to reduce risk of uncontrolled release of formation fluids throughout the life cycle of a well”. Also, “There shall be two well barriers available during all well activities and operations, including suspended or abandoned wells, where a pressure differential exists that may cause uncontrolled outflow from the borehole/well to the external environment.” The first part of D-010 is the same as API 100-1. The D-010 requirement for two barriers at all times is not present in API; however, the BSEE Final Drilling Rule for offshore drilling does require this when removing well control equipment. Note API 100-1 applies to onshore wells, API RP 65-2 (API, 2010) and API 96 (API, 2013a) address offshore well integrity (and some use for onshore wells), along with several other API standards. BSEE (2016) specifies how to achieve we ll integrity in its Final Drilling Rule, including all necessary permits. This rule does not apply to onshore wells which follow API standards, such as API 100-1. Conventional (basic) representation Detailed re presentation

204 INVESTIGATING PROCESS SAFETY INCIDENTS Root Cause— A fundamental, underlying, system- related reason why an incident occurred that identifies a correctable failure(s) in management systems. There is typically more than one root cause for every process safety incident. Correcting only a causal factor is a simplistic approach that may prevent the identical incident from occurring again at the same location, but will not prevent similar incidents. Identifying and correcting the root causes should eliminate or substantially reduce the likelihood of recurrence of the incident and other similar incidents at the location. More importantly, the new knowledge and corrective methods resulting from the investigation may be shared for use throughout a company and possibly apply to an industry as a whole. A thorough incident investigation i dentifies and addresses all of the causes of an incident, including the root causes. It also provides the mechanism for under standing the interaction and impact of management system failures. This analysis provi des the means for fu lly addressing the incident, similar inciden ts, and even dissimilar incidents caused by the same root causes, throughout the facility, company, and industry. Addressing management system failures is the ultimate goal, yielding the maximum benefit from an inci dent investigation. The following example illustrates the concept of root cause analysis. Consider a scenario where a worker steps into a puddle of oil on the plant floor, slips, and falls. A traditional investigation might identify “oil spilled on the floor” as the cause, with the remedy limited to cleaning up this particular spill and possibly admonishing the worker for not being more careful. By using the tools described in this chapter, it will be clear that the oil on the floor is actually a symptom of underlying causes, rather than a root cause of itself. A structured root cause investig ation explores the underlying causes and examines the systems and condit ions involved in the incident. . . . It is from identifying the underlying causes that the most benefit is gained. By addressing only the causal factor, the identical accident is prevented from occurring again; by addressing the underlying root cause(s), numerous other similar incidents are prevented from occurring. . .

10 • Risk Based Process Safety Considerations 204 operational discipline (discussed next), better design and implementation of effective process safety systems, and improved process safety performance. 10.4 Effects of weak operational discipline One of the foundations of an effective process safety program, closely linked to its commitment to process safety pillar and to sustaining its management systems in Pillars II, III, and IV, is the organization’s Operational Discipline (OD). A weak Conduct of Operations element (Pillar III, Element 15) is reflected by weaknesses in the company’s OD. Although OD is difficult to measure, it is useful to think of its impact on risk by expressing its qualitative effect using the following simplified risk equation: Details on this equation are provided in other publications [21, p. 85] [49]. As an example of how OD qualitatively affects the risk, the OD term is in the denominator: The values of the denominator of Equation 10.1 indicate the following: 1/1 (or “1”) representing 100% OD, where everyone does everything right every time, or 1/2 (or “0.5”) representing 50% OD, where everything is done correctly—or incorrectly— only half of the time . Thus, the actual risk at 50% OD is twice the expected risk with 100% OD. As OD performance increases, the closer it approaches 100% compliance and effective conduct of operations for everyone in the

462

Plant Interlocks and Alarms 335 A HIPPS is a type of SIS that cuts off a high pressure, rebel stream to stop it from getting into a non‐ or par - tially protected system (Figure 16.4). Implementing a HIPPS is highly regulated because it may release the need of pressure safety device down- stream of HIPPS. HIPPSs are used to get rid of a pressure safety device or decrease its size. A HIPPS is used where one or more pressure safety devices has such large release rate that severely impacts the collection network and/or emergent release disposal system. The other cases in which a– HIPPS may be needed is for the cases where putting a pressure safety valve is not technically doable, like under‐the‐sea facilities. The result of implementing a HIPPS on the downstr eam system could be reduced size of PSV, removing PSV, or reducing the design pressure of the downstream system. A BMS is a type of SIS that most likely exists when dealing with burners. Burners could be in a fired heater, boiler, steam generator or even a tunnel dryer. A BMS is a regulated practice in many countries. Wherever there is a burner in the system, a BMS needs to be implemented. A BMS includes actions similar to ones in an ESD and emergency isolation plus additional actions to push the potential accumulated flammable mixture out of the sy stem. The BMS actions could be initiating snuffing steam and/or compressed air into the firing box. A BMS could be a combination of other SISs like an ESD, emergency isolation, etc. BMS is a collective name for different practices, but just few of them are visible on P&IDs.Y 218I/P Block valves Bleed valveY 220I/PY 219 To-be-isolated syste mI/P Figure 16.3 Aut omatic double block and vent/bleed switching valves. PT 124PT 125SD SD Fully protected sys tem by mechanical safety deviceNon- or partiall y- protected syste m by mechanical safety devicePT 123Figure 16.4 A typical HIIPS.

218 time observation, interception and post -incident analysis of the activities and identity of the adversary. Delay : A security strategy to provid e various barriers to slow the progress of an adversary in penetratin g a site to prevent an attack or theft, or in leaving a restricted ar ea to assist in apprehension and prevention of theft. Respond : The act of reacting to detect ed criminal activity either immediately following detection, such as notifying local authorities for assistance, or post-incident via surveillance tapes or logs. Resilience/Resiliency : (Ref 9.8 DHS) The ability to resist, absorb, recover from, or successfully adap t to adversity or a change in conditions. In the context of energy security, resilience is measured in terms of robustness, resource fulness, and rapid recovery. A complete security design includes these concepts in “Layers of Protection” or a “Defense in Depth” arrangement (Figure 9.1). Ideally, the most critical assets should be in th e center of the conceptual concentric levels of increasingly more stri ngent security measures. Security scenarios often include direct attack s on or near an asset. For this reason, the spatial relationship or pr oximity between the location of the target asset and the location of the physical countermeasures is important. The facility could detect, deter, delay, or respond to the event at multiple points in the layers of se curity. Optimally, security would deter and detect at the outermost layers to provide sufficient time for responders to suppress or neutralize th e adversary prior to initiation of the event. The layers of protection for critical assets may need to be quite robust because the adversaries are intentionally attempting to breach the protective features and may use whatever means are available to help ensure a successful attack. This may include the use of explosives or other destructive events that re sult in widespread common cause failures. Particularly motivated adversaries may go to extreme lengths, including suicide attacks, to breach the security layers of protection.

30 | 3 Obstacles to Learning capsule design. The original design featured an oxygen-rich atmosphere, lots of nylon straps, and an inward-opening hatch. During the test, an electrical short occurred. Nylon normally smolders in air, but the straps combusted rapidly in the capsule’s high-oxygen atmosphere. Once the fire started, the pressure from the combustion gases tightly sealed the hatch. Despite the efforts of the crew inside the capsule and ground support staff outside, the hatch could not be opened. It was impossible for the astronauts to escape. The investigation committee called this breakdown of the hazard analysis a failure of imagination. However, the astronauts themselves imagined it. During a planning meeting, they asked that flammables be removed from the Apollo cabin. But the designers gave this step a low priority, partly because NASA considered a cabin fire improbable. The astronauts also were said to have asked for an outward-opening door. The Apollo 1 fire is an example of the natural human tendency to understate potential consequences, understate probability of occurrence, and overstate the effectiveness of preventive and mitigative barriers. These tendencies become exacerbated when hazard and risk analysis is performed under time pressure with competing priorities. A lesson learned that became institutional knowledge In May 2020, SpaceX and NASA successfully launched the Crew Dragon spacecraft to the International Space Station. At the pre-launch briefing, Astronaut Doug Hurley said, “On more than one occasion he [Elon Musk] has looked both Bob [Astronaut Behnken] and me right in the eye and said, 'Hey, if there's anything you guys are not comfortable with or that you're seeing, please tell me and we'll fix it.’” (Grush 2020). This appears to be a step in the right direction for institutionalizing past lessons learned. Now NASA and its suppliers need to ensure this lesson remains institutionalized. Lack of Understanding About Hazards Sometimes, you don’t know to look for a hazard until the hazard finds you. The 1960 aniline plant explosion in Kingsport, TN, is one good example. In the presentation “Let Me Tell You…The Impact of Eastman’s Aniline Plant Explosion on Process Safety Awareness” at the 11th Global Congress on Process Safety (Lodal 2015), the speaker noted that at the time of the 1960 investigation, the cause of the explosive reaction was unknown. To this day, the cause is still only speculative, but it has spurred on greater awareness that process safety hazards can exist in unexpected places.

Part 1: Concepts, principles, and foundational knowledge Human Factors Handbook For Process Plant Operations: Improving Process Safety and System Performance CCPS. © 2022 CCPS. Published 2022 The American Institute of Chemical Engineers.

120 GUIDELINES FOR MANAGING ABNORMAL SITUATIONS Figure 5.1 Protection and Their Impact on the Process Unlike other safeguards or layers of protection, such as a pressure relief valve or safety instrumented system (SIS), the operator’s response to an alarm relies on human interv ention. There are numerous potential failure modes for operator response to an alarm including hardware, software, and human behavior. Failu res in human behavior become more likely with poor alarm system design and performance (nuisance alarms, stale alarms, redundant alarms, and alarm floods). These failures are often improperly labeled as “operator error”; but are often more appropriately characterized as alarm management failures. In summary, alarm management and control panel operators are critical layers of safeguard protec tion. However, when it comes to management of abnormal situations in the plant, traditional distributed control system (DCS) alarms are of ten not enough. Ma nufacturers have developed methods to apply boundary management tools that test and manage the process limits. To work effectively, however, a variety of structured and unstructured process in put data must be aggregated in a common database for continual an alysis. Only then can predictive analytics and root cause analysis be proactively applied to prevent processes from drifting into abnorm al situations or unwanted events.

xxix PREFACE The project to produce this concept bo ok was a joint collaboration between CCPS and the Society of Petroleum Engineers (SPE). The Center for Chemical Process Safety (CCPS) was established in 1985 to protect people, property and the environment from major chemical incidents by bringing best practices and knowledge to industry, academia, the government and the public around the world. As part of this vision, CCPS has focused on developing and disseminating technical information through collective wisdom, tools, training and expertise from experts within the oil, gas, and petrochemical industry. The primary source of this inform ation is a series of guideline and concept books to assist industry in implementing various elements of process safety and risk management. This concept book is part of this series. SPE is a transnational technical and professional society serving members engaged in the exploration, development, production and mid-stream segments of the oil, gas, and related industries. It has a mission to collect, disseminate, and exchange technical knowledge concerning the exploration, development and production of oil and gas resources and related technologies for the public benefit. As a not-for-profit organization, CCPS has published over 100 books, written by member company representatives who have donated their time, talents and knowledge. Industry experts, and contract ors that prepare th e books, typically provide their services at a discount in ex change for the recognition received for their contributions in preparing these books for publication.

75 4.4 REFRIGERANTS Refrigeration systems require the use of a material that has a high vapor pressure, low flash point, and high he at capacity in order to function successfully as a coolin g medium. Unfortunately, most of the materials that have desirable properties as refrigerants also have other properties that introduce undesirable risks. The three major refrigerants used in industrial refrigeration systems are light hydrocarbons (i.e., ethylene, propane, propylene, etc.), anhydrous ammonia, and chlorofluorocarbons (CFC). Each of th ese materials presents individual hazards that may preclude their use in refrigeration systems. Hydrocarbons, such as propane and propylene, are highly flammable and represent significant fire and ex plosion hazards if released due to leakage. Anhydrous ammonia is a toxi c inhalation hazard, and CFCs, first developed in the 1920’s as a sa fer alternative to ammonia and hydrocarbon-based refrigerants, deplete the earth’s protective ozone layer when released to the atmosphere. The choice of refrigerant will depend on availability of the material on-site, ease of maintenance of the refrigeration system, and amount of refrigeration tonnage required for the application. Due to the intern ational banning of CFC usage, many facilities have changed back to the us e of hydrocarbon refrigerants. If a hydrocarbon-based refrigeration syst em only incremen tally increases the total amount of flammable hydrocar bons on-site, it may be the best choice. Anhydrous ammonia refriger ation systems have been used extensively in the food processing industry because ammonia is an environmentally-compatible refrigerant with no ozone depletion potential (ODP). Ammonia also has advantageous thermodynamic properties at the temperatures and pressures common to refrigeration applications, resulting in smaller, more compact systems and less energy consumption than other refrigerants when used in large industrial systems. There has been research in recent years on alternative refrigeration materials that have lower ODP levels , or lower flammability, such as hydrofluorocarbons, hydrochlorofluorocarbons (HCFC), and hydrofluoroethers. These materials can be used on their own or as carriers for flammable materials, result ing in a refrigerant that is less flammable than a pure hydrocarbon, su ch as propane (i.e., substitution). HCFCs are scheduled for phase-out in the first third of the 21st century

110 considered. These factors can sign ificantly increase the maximum pressure of a combustion reaction. In summary, when robust equipment design virtually eliminates equipment failure, it can be consid ered an inherently safer design. Therefore, it fits within the definiti on of simplification. It is highly effective in eliminatin g the possibility of an uncontrolled loss of containment. In a general sense, the removal of this possibility from a process design must be inherently safer. 6.4 PREVENTING RUNAWAY REACTIONS Choosing the addition order of reac tants to mitigate or eliminate a potential runaway reaction can redu ce the hazard associated with a process and may allow for a simplified emergency relief system. It is essential that the reaction mechan isms, thermodynamics, and kinetics under runaway conditions be thoroughly understood to be confident that the design pressure is sufficie ntly high for all credible reaction scenarios. All causes of a runaway reaction must be understood, and any side reactions, decompositions, and shifts in reaction paths at the elevated temperatures and pressu res experienced under runaway conditions must be evaluated. Ma ny laboratory test devices and procedures are available for evalua ting the consequences of runaway reactions (Ref 6.3 CCPS 1995a; Ref 6.4 CCPS 1995b). Several of these reaction hazard testing methods are summarized in Table 6.1 (Ref 6.5, CCPS 2003). Table 6.1 Summary of Reaction Testing Methods (Ref. 6.5 CCPS 2003) Hazards Test Stage Method Typical Information Comments Hazard Screening Desk Calculation Reaction enthalpy ΔH RXN Need formation energy data or must derive it. Must know precise stoichiometry

274 Table 11.1 (Ref 11.16 Kletz 2010). Others have been published in the literature (Ref 11.7 Bollinger) (Ref 11.10 CCPS 1998). A more extensive version is included in Appendix A of this book. In addition to the general use of IS checklists in PH As/HIRAs, reviews devoted to only IS are sometimes used where the four IS strategies are examined explicitly to determine if there are feasible ways to use them. Such IS-only reviews are currently required in two United States jurisdictions: Contra Costa County, California, and New Jersey (See Chapter 14). Often the format and study approach for these IS-only reviews is PHA-like, i.e., worksheets, guidewords/deviations, and similar techniques are often used for cond ucting them. Contra Costa Health Services, in their IS guidance document includes analytical activities on how IS should be addressed for ne w vs. existing processes (Ref 14.13 CCHS). See Chapter 14. Amyotte and Kletz also state that PH As for capital project reviews, as well as other PHAs that are perfor med should include the following purposes: to identify hazards, hazardous situ ations and specific events that could produce undesirable consequences, to examine the currently available safety measures to deal with the identified hazards and events, and to suggest alternatives for risk reduction based on inherent, as well as engineered (add-on) and procedural, safety. (Ref 11.16 Kletz 2010) In Appendix B of this book, an extensive example of IS-specific reviews using a PHA-like approach is presented. Facility siting is usually considered part of the PHA/HIRA element of PSM. Facility siting is examined qu alitatively during PHAs/HIRAs using checklists typically. It is also addressed quantitative ly in most facilities as a separate analytical activity. The vulnerability to the effects of explosions of the administration an d control buildings at Flixborough and the contractor trailers at the BP Texas City refinery are often cited as examples of the importance of th is topic. Additionally, the use of segregation in the hierarchy of controls and the use of distance in facility siting are examples of Moderation .

| 203 6 WHERE DO YOU START? 6.1 IN TRODUCTION Evaluating and then m odifying the process safety culture of your facility or com pany can be a daunting venture, particularly if the required culture change is significant. This chapter addresses how to get started and provides a roadmap for your culture journey. First, you should acknowledge that there may be existing or developing weaknesses in the process safety culture. Even the com panies with the best performance in process safety have som e weaknesses or have the potential to develop them. A com pany that denies that it m ay have weaknesses in its process safety culture has at least one weakness – a decreased sense of vulnerability – and probably other weaknesses as well . In stronger cultures, the feeling that weaknesses have all been corrected indicates a sense of complacency that can quickly compound via normalization of deviance . In less-developed cultures, denial m ay be based on a false sense of security taken from the wrong m etrics or a focus on compliance, indicating a weak imperative for process safety . Therefore, all com panies should search for cultural weaknesses, regardless of where they are on their process safety journey. Making the case for culture change can be challenging. Marshall the facts carefully to show how process safety supports business and financial success, and how improving process safety Essential Practices for Creating, Strengthening , and Sustaining Process Safety Culture , First Edition. CCPS . © 2018 AIChE . Published 2018 by John Wiley & Sons, Inc .

HUM AN FACTORS 267 processing systems fail to consider reasonable human capability limits and patterns of habit. The result can often be a system that promotes human errors rather than discouraging them. Donald Norman addresses these mismatches comprehensively in the book The Design of Everyday Things (Norman, 1988). Human performance problems occur several ways. Reason outlined several types of involuntary or uninte ntional human actions (Reason, 1990). The Skills, Rules, Knowledge (SRK) model was developed by Rasmussen (Rasmussen, 1983) to help designers combine information requirements for a system and aspects of human cognition. As an investigator uses tools such as 5 Whys to identify potential root causes, considering these models can help focus in on specific areas for improvement to support the desired human performance. 11.2 INCORPORATING HUM AN FACTORS INTO THE INCIDENT INVESTIGATION PROCESS As stated at the begi nning of this chapter, humans are involved in all aspects of the workplace. In addition to managing, designing, operating, and maintaining, this also includes investigation and learning. Thus, nurturing a blame-free, open culture within an organization is essential for the success of the incident investigation process. The investigation must focus on understanding: • What happened? • How did it happen? • Why did it happen? • What can be done to preven t it from happening again? • How can the risk be reduced? There are a number of references sp ecifically addressing human factors as related to incident investigation that the reader may find useful. Two of note are the Energy Institute’s “Learn ing from incident, accidents and events” (EI, 2016) and the International Association of Oil & Gas Producers’ “Demystifying Human Factors: Build ing confidence in human factors investigation” (IOGP, 2018).

84 PROCESS SAFETY IN UPSTREAM OIL & GAS 4.3.6 Learn from Experience Learn from Experience is one of the 4 pillars of RBPS. This pillar includes Incident Investigation, Measurement and Metrics, Auditing, and Management Review and Continual Improvement . These tools are also in API RP 75 SEMS which is required by regulation in the US OCS. CCPS (2008b and 2019c) has published texts on incidents that define process safety. While this is very useful for periodic safety talks and to build process safety knowledge in ne wer personnel, this primarily focuses on downstream incidents, although it does include the Deepwater Horizon incident. Effective hazard identification requires that teams identify all potential process safety events. Most companies have in ternal communication systems to share incident lessons. However, process safety incidents are rare and may not have occurred within the facility’s experience. There are at least four important mechan isms to help companies ensure that personnel are aware of these rare yet important incidents. 1.Regular reviews of industry incidents – In the US, state regulators report on onshore drilling incidents, and the CSB (2019) has started to investigate onshore drilling incidents. BSEE and other offshore international regulators also report major incidents both on their own websites as well as a consolidated list on the International Regulators Forum website. API RP 754, and its offshore equivalent IOGP 456, provide definitions of leading and lagging indicators for four tiers of process safety events. Companies are starting to follow these guidelines and are reporting publicly on the more serious Tier 1 and 2 events. COS provides a listing of incidents. Further details were provided in Chapter 1. It is the role of process safety specialists and design engineers in the company to monitor such statistics and provide safety talks or updated company design rules to address such incidents. 2.Participate in and employ current codes and standards – Engineering bodies (e.g., API, ANSI, IADC) update their codes and standards periodically and these updates address any important incidents if the existing documents do not address the issues adequately. Companies that participate in these committees get early notice of changes and, by interacting with other company specialists in th e committee, learn of inci dents that may not be published. 3.Participate in industry co nferences and public meetings – CCPS, SPE, API, COS, IADC and other bodies organize periodic conferences which address technical advances, upcoming standard updates, and often provide recent incident summaries. Participating in these events helps process safety and well construction specialists keep up to date on technology and aware of industry incidents. 4.Training in process safety – Process safety training is available from industry associations including CCPS. The CCPS offers Safety and Chemical Engineering Education (SAChE) courses focusing on university students addressing process safety and the RBPS system (SAChE, 2019).